Eye Tracking Method and User Terminal Performing Same

1. A user terminal comprising: an imaging device configured to capture a face image of a user; and at least one processor configured to execute program codes to perform: acquiring a vector representing a direction that a face of the user faces and an ocular image of the user from the face image on the basis of set rules; and tracking a gaze of the user by inputting the face image, the vector, and the ocular image to a set deep learning model, wherein the deep learning model has been trained with training data including a face image of a viewer and location information of a set point which the viewer looks at in a screen; the training data is collected at a time point at which the viewer touches the point or the viewer looks the point or an utterance of the viewer related to text displayed at the point is started, and wherein the imaging device is maintained in an off state and operated to photograph the viewer at a time point at which the viewer touches the point.

2. The user terminal of claim 1 , wherein the tracking of the gaze of the user comprises tracking the gaze of the user using the deep learning model which has learned the training data.

3. The user terminal of claim 1 , wherein the training data is collected by the imaging device operating at the time point at which the viewer touches the point.

4. The user terminal of claim 1 , wherein the at least one processor is further configured to execute program to perform transmitting the training data collected at the time point at which the viewer touches the point to a server.

5. The user terminal of claim 1 , wherein, when the viewer touches the point while the imaging device is operating, the training data is separately collected at the time point at which the touch is made and time points a set time before and after the time point at which the touch is made.

6. The user terminal of claim 1 , wherein the at least one processor is further configured to execute program to perform changing a visual element of the point after the viewer touches the point so that a gaze of the viewer remains at the point even after the touch.

7. The user terminal of claim 1 , wherein the at least one processor is further configured to execute program to perform acquiring ocular location coordinates and face location coordinates of the user from the face image on the basis of the rules; the tracking of the gaze of the user comprises additionally inputting the ocular location coordinates and the face location coordinates to the deep learning model together with the vector representing the direction that the face of the user faces.

8. The user terminal of claim 1 , wherein the at least one processor is further configured to execute program to perform: displaying advertising content on the screen, determining whether the user is watching the advertising content on the basis of a detected gaze of the user and a location of the advertising content in the screen; and changing the location of the advertising content in the screen by considering the location of the advertising content in the screen and a time period for which the user has watched the advertising content.

9. An eye tracking method comprising: capturing, by an imaging device, a face image of a user; acquiring, by an eye tracking unit, a vector representing a direction in that a face of the user faces and an ocular image of the user from the face image on the basis of set rules; and inputting, by the eye tracking unit, the face image, the vector, and the ocular image to a set deep learning model to track a gaze of the user, wherein the deep learning model has been trained with training data including a face image of a viewer and location information of a set point which the viewer looks at in a screen; the training data is collected at a time point at which the viewer touches the point or the viewer looks the point or an utterance of the viewer related to text displayed at the point is started; and wherein the imaging device is maintained in an off state and operated to photograph the viewer at a time point at which the viewer touches the point.

10. The eye tracking method of claim 9 , wherein the tracking of the gaze of the user comprises tracking the gaze of the user by using the deep learning model which has learned the training data.

11. The eye tracking method of claim 9 , wherein the collecting of the training data comprises collecting the training data by operating the imaging device at the time point at which the viewer touches the point.

12. The eye tracking method of claim 9 , further comprising transmitting, by the training data collection unit, the training data collected at the time point at which the viewer touches the point to a server.

13. The eye tracking method of claim 9 , wherein the collecting of the training data comprises, when the viewer touches the point while the imaging device is operating, separately collecting the training data at the time point at which the touch is made and time points a set time before and after the time point at which the touch is made.

14. The eye tracking method of claim 9 , further comprising changing, by the training data collection unit, a visual element of the point after the viewer touches the point so that a gaze of the viewer remains at the point even after the touch.

15. The eye tracking method of claim 9 , further comprising acquiring, by the eye tracking unit, ocular location coordinates and face location coordinates of the user from the face image on the basis of the rules, wherein the tracking of the gaze of the user comprises additionally inputting the ocular location coordinates and the face location coordinates to the deep learning model together with the vector representing the direction that the face of the user faces.

16. The eye tracking method of claim 9 , further comprising: displaying, by a content providing unit, advertising content on the screen; determining, by the eye tracking unit, whether the user is watching the advertising content on the basis of a detected gaze of the user and a location of the advertising content in the screen; and changing, by the content providing unit, the location of the advertising content in the screen by considering the location of the advertising content in the screen and a time period for which the user has watched the advertising content.